INDEX

Explanations

terms and phrases related to social and legal equality issues

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__mlp-sm_processed/2-mlp-sm

Prompts (Dashboard)

32,768 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

torch.float32

Hook Name

blocks.2.hook_mlp_out

Hook Layer

Architecture

standard

Context Size

128

Dataset

EleutherAI/the_pile_deduplicated

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

amiento

-1.51

ylum

-1.50

ement

-1.47

 yourselves

-1.40

iors

-1.40

nement

-1.39

nem

-1.37

]:

-1.35

 generality

-1.34

Suite

-1.32

POSITIVE LOGITS

 Finally

1.48

 followed

1.46

ķ

1.43

 simultaneous

1.41

´

1.41

 goose

1.35

tac

1.35

 knowingly

1.33

↵↵ ↵

1.30

Finally

1.27

Activations Density 2.926%

terms and phrases related to social and legal equality issues

No Comments

No Known Activations